The goal of the proposed research is to better understand the nature of pragmatic impairments in autism by employing experimental methods from psycholinguistics. This application aims to clarify what types of contextual information are considered in the real-time language processing of individuals with autism. By doing so, the predictions of two accounts of autism, a primary deficit in Theory of Mind (Baron-Cohen, 1995) and Weak Central Coherence theory (Frith, 1989) can be evaluated. ToM predicts that individuals with autism should have difficulties using language when this depends on understanding another's perspective and communicative intent, while WCC predicts that they will have difficulty when any type of contextual information is important for interpretation. A second aim is to examine real-time language processing in individuals with autism, by the use of eye movement methodology to analyze participants' eye movements during spoken language interpretation. These studies will offer a window into reasoning processes in autism in general, and empirical evaluation of the predictions of ToM and WCC for the domain of pragmatics. This research may also provide a basis for effective communicative strategies for individuals with autism.